6:51 a.m.
I'm running r10651 on RO 5.16 (flashed) with !HID 0.36. I've noticed
recently that if I use the scroll bar to move up and down a page all
is OK, but if I try the scroll wheel (which is my normal method) I get
a hang of 6 or 7 seconds each time before the page moves. During this
time the Organiser clock stops. The worst case was around 30 seconds,
but I've not seen it that long since. Can anyone else replicate this?
--
Grahame Parish
7:12 a.m.
On 19 Jul 2010 Grahame Parish <maillist.parish(a)millers-way.net>
wrote:
I'm running r10651 on RO 5.16 (flashed) with !HID 0.36. I've
noticed
recently that if I use the scroll bar to move up and down a page all
is OK, but if I try the scroll wheel (which is my normal method) I get
a hang of 6 or 7 seconds each time before the page moves. During this
time the Organiser clock stops. The worst case was around 30 seconds,
but I've not seen it that long since. Can anyone else replicate this?
Yes, same OS, build 10654 but without !HID (I could never get my head
around this). Slightly different here, though; the first page visited
gives only a slight delay, less than a second. With subsequent pages
the delay gets worse, up to the 30 seconds or so, with Organizer's
clock hanging, on about the fourth page.
I've just reverted to 10649, where this doesn't happen. Odd! Will you
file a bug report?
With best wishes,
Peter.
--
Peter, \ / zfc Lu \ Prestbury, Cheltenham, Glos. GL52
Anne \/ __ __ \ England.
and / / \ | | |\ | / _ \ http://pnyoung.orpheusweb.co.uk
family / \__/ \_/ | \| \__/ \______________ pnyoung(a)ormail.co.uk
8:01 a.m.
In message <54583d3951.pnyoung(a)pnyoung.ormail.co.uk>
Dr Peter Young <pnyoung(a)ormail.co.uk> wrote:
On 19 Jul 2010 Grahame Parish
<maillist.parish(a)millers-way.net>
wrote:
> I'm running r10651 on RO 5.16 (flashed) with !HID 0.36.
I've noticed
> recently that if I use the scroll bar to move up and down a page all
> is OK, but if I try the scroll wheel (which is my normal method) I get
> a hang of 6 or 7 seconds each time before the page moves. During this
> time the Organiser clock stops. The worst case was around 30 seconds,
> but I've not seen it that long since. Can anyone else replicate this?
Yes, same OS, build 10654 but without !HID (I could never get my
head
around this). Slightly different here, though; the first page visited
gives only a slight delay, less than a second. With subsequent pages
the delay gets worse, up to the 30 seconds or so, with Organizer's
clock hanging, on about the fourth page.
I've just reverted to 10649, where this doesn't happen. Odd!
Will you
file a bug report?
With best wishes,
Peter.
Bug raised - although I forgot to put in version numbers and can't see
how to edit the submission. Have tested with 10654 and it still
occurs, but doesn't seem to freeze for as long on tests so far.
Cheers,
Grahame.
--
Grahame Parish
8:50 a.m.
In article <54583d3951.pnyoung(a)pnyoung.ormail.co.uk>,
Dr Peter Young <pnyoung(a)ormail.co.uk> wrote:
Yes, same OS, build 10654 but without !HID (I could never get my head
around this). Slightly different here, though; the first page visited
gives only a slight delay, less than a second. With subsequent pages
the delay gets worse, up to the 30 seconds or so, with Organizer's
clock hanging, on about the fourth page.
Do you still get the scroll delays while NetSurf is not fetching anything?
I've just reverted to 10649, where this doesn't happen. Odd!
Will you
file a bug report?
The changes between these revisions was to add a lot of extra fetch and
cache logging. We've had bug reports of crashes in the fetch layer, which
we've been unable to diagnose and fix. (e.g. the recent "NetSurf failure
on a BBC site" thread.) The additional logging has been added in the hope
that someone can report the same bug, this time with far more
comprehensive information in the log file.
--
Michael Drake (tlsa) http://www.netsurf-browser.org/
9:29 a.m.
In message <5139465479tlsa(a)netsurf-browser.org>
Michael Drake <tlsa(a)netsurf-browser.org> wrote:
In article <54583d3951.pnyoung(a)pnyoung.ormail.co.uk>,
Dr Peter Young <pnyoung(a)ormail.co.uk> wrote:
> Yes, same OS, build 10654 but without !HID (I could never get my
head
> around this). Slightly different here, though; the first page visited
> gives only a slight delay, less than a second. With subsequent pages
> the delay gets worse, up to the 30 seconds or so, with Organizer's
> clock hanging, on about the fourth page.
Do you still get the scroll delays while NetSurf is not fetching
anything?
I usually wait for the load to complete before scrolling. I've
noticed that testing so far with 10654 the hangs are down to a few
seconds for each scroll attempt. Most of my testing has been on the
ROOL site http://www.riscosopen.org/content/ and the NetSurf latest
test builds page (I use !RiscPkg to update, but I read the test builds
page to see if there's been anything new that affects the RISC OS
version). I use use the default window size initially, but often
maximise the window if there's more than a small amount to read up.
NetSurf is generally running for days at a time except when updating
it and the Iyonix runs 24/7.
> I've just reverted to 10649, where this doesn't happen.
Odd! Will you
> file a bug report?
The changes between these revisions was to add a lot of extra fetch
and
cache logging. We've had bug reports of crashes in the fetch layer, which
we've been unable to diagnose and fix. (e.g. the recent "NetSurf failure
on a BBC site" thread.) The additional logging has been added in the hope
that someone can report the same bug, this time with far more
comprehensive information in the log file.
I've not experienced the reported failures, and since the layout was
fixed on the new BBC news site I've had no problems there at all.
Cheers,
Grahame.
--
Grahame Parish
12:57 p.m.
I don't know if it's useful (or even relevant) but I've loaded
!TaskUsage to see what is happening in terms of processor use by
NetSurf. Here's the results (RO 5.16, NS r10654): -
NetSurf loaded on iconbar, no window open: 0%
NetSurf homepage fully loaded and displayed without having focus: 0%
NetSurf homepage fully loaded and displayed with focus, without
pointer: 0%
NetSurf homepage fully loaded and displayed with focus, with pointer
over window: 57-61%
NetSurf on ROOL Forum list page <http://www.riscosopen.org/forum/>;,
scrolled top to bottom with scroll bar: peak 45%
NetSurf on ROOL Forum list page, scrolled top to bottom with scroll
wheel: 93-100% per scroll wheel movement.
--
Grahame Parish
5:35 a.m.
A further observation - now running 10657. When first loaded NetSurf
responds normally to the scroll wheel, but after a viewing as few
simple pages and scrolling them it exhibits the same hanging problem.
I upgraded to 10657, ran NetSurf, went to ROOL and scrolled up and
down the ROOL homepage without problems several times. Went to
riscos.info home page and after a few scrolls the response got worse.
Not sure if this is time-related, activity-related or something else.
Would a log file help at all?
--
Grahame Parish
5:52 a.m.
On 23 Jul 2010 Grahame Parish wrote:
A further observation - now running 10657. When first loaded
NetSurf
responds normally to the scroll wheel, but after a viewing as few
simple pages and scrolling them it exhibits the same hanging problem.
I upgraded to 10657, ran NetSurf, went to ROOL and scrolled up and
down the ROOL homepage without problems several times. Went to
riscos.info home page and after a few scrolls the response got worse.
Not sure if this is time-related, activity-related or something
else.
Would a log file help at all?
Netsurf has always been liable to keel over if you scroll while it is
loading a page. This may be content related - my sites are generally
OK but I am very wary of scrolling with the scroll wheel or scroll
bars, or resizing the window, while a page is loading.
--
Richard Porter http://www.minijem.plus.com/
mailto:ricp@minijem.plus.com
Confidentiality statement: Imagine 50 lines of pointless verbiage.
6:08 a.m.
In article <1271453b51.ricp(a)user.minijem.plus.com>,
Richard Porter <ricp(a)minijem.plus.com> wrote:
NetSurf has always been liable to keel over if you scroll while it is
loading a page. This may be content related - my sites are generally
OK but I am very wary of scrolling with the scroll wheel or scroll
bars, or resizing the window, while a page is loading.
Really? I do these things routinely.
A crash log for such an incident would be helpful.
--
Michael Drake (tlsa) http://www.netsurf-browser.org/
6:31 a.m.
On 23 Jul 2010 Michael Drake wrote:
In article <1271453b51.ricp(a)user.minijem.plus.com>,
Richard Porter <ricp(a)minijem.plus.com> wrote:
> NetSurf has always been liable to keel over if you scroll while
it is
> loading a page. This may be content related - my sites are generally
> OK but I am very wary of scrolling with the scroll wheel or scroll
> bars, or resizing the window, while a page is loading.
Really? I do these things routinely.
A crash log for such an incident would be helpful.
I generally do post the log if it pops up. If I have to reset the
computer than I'll probably forget to dig it out before running
NetSurf again. The bug will usually be reported as a segmentation
fault or abort as appropriate.
--
Richard Porter http://www.minijem.plus.com/
mailto:ricp@minijem.plus.com
Confidentiality statement: Imagine 50 lines of pointless verbiage.
6:39 a.m.
In article <3cf9483b51.ricp(a)user.minijem.plus.com>,
Richard Porter <ricp(a)minijem.plus.com> wrote:
I generally do post the log if it pops up. If I have to reset the
computer than I'll probably forget to dig it out before running
NetSurf again. The bug will usually be reported as a segmentation
fault or abort as appropriate.
A log from a recent version with all the extra fetch / cache logging would
be useful, if it happens while things are still fetching. So r10650 or
later.
--
Michael Drake (tlsa) http://www.netsurf-browser.org/
11:42 a.m.
On 23 Jul, Richard Porter wrote in message
<3cf9483b51.ricp(a)user.minijem.plus.com>:
I generally do post the log if it pops up. If I have to reset the
computer
than I'll probably forget to dig it out before running NetSurf again. The
bug will usually be reported as a segmentation fault or abort as
appropriate.
Where is the crash coming from (NetSurf, or the scrolling module), on what
OS, and with what mouse support utilities (if any)? HID is known to crash
out on scrolling with certain types of window, for example.
--
Steve Fryatt - Leeds, England
http://www.stevefryatt.org.uk/
12:01 p.m.
On 23 Jul 2010 Steve Fryatt wrote:
Where is the crash coming from (NetSurf, or the scrolling module),
on
what OS, and with what mouse support utilities (if any)? HID is known
to crash out on scrolling with certain types of window, for example.
OS 6.16, Mercury PS2 mouse with Mouse Mini. I don't know where the
crash occurs.
--
Richard Porter http://www.minijem.plus.com/
mailto:ricp@minijem.plus.com
Confidentiality statement: Imagine 50 lines of pointless verbiage.
6:20 a.m.
In message <1271453b51.ricp(a)user.minijem.plus.com>
Richard Porter <ricp(a)minijem.plus.com> wrote:
On 23 Jul 2010 Grahame Parish wrote:
> A further observation - now running 10657. When first loaded
NetSurf
> responds normally to the scroll wheel, but after a viewing as few
> simple pages and scrolling them it exhibits the same hanging problem.
> I upgraded to 10657, ran NetSurf, went to ROOL and scrolled up
and
> down the ROOL homepage without problems several times. Went to
> riscos.info home page and after a few scrolls the response got worse.
> Not sure if this is time-related, activity-related or something
else.
> Would a log file help at all?
Netsurf has always been liable to keel over if you scroll while it
is
loading a page. This may be content related - my sites are generally
OK but I am very wary of scrolling with the scroll wheel or scroll
bars, or resizing the window, while a page is loading.
All of the problems reported have occurred with fully loaded pages. I
very rarely scroll before the page has completely loaded.
--
Grahame Parish
6:53 a.m.
In message <01ec473b51.GrahameParish(a)grahame.parish>
Grahame Parish <maillist.parish(a)millers-way.net> wrote:
All of the problems reported have occurred with fully loaded pages.
I
very rarely scroll before the page has completely loaded.
Further note: When NetSurf is first loaded I can scroll the first
page or two viewed with the scroll wheel without any problems. After
that it starts hanging and only moving one 'step' no matter how much I
scroll each time.
--
Grahame Parish
6:57 p.m.
In message <5fe1e53e51.GrahameParish(a)grahame.parish>
Grahame Parish <maillist.parish(a)millers-way.net> wrote:
In message <01ec473b51.GrahameParish(a)grahame.parish>
Grahame Parish <maillist.parish(a)millers-way.net> wrote:
> All of the problems reported have occurred with fully loaded
pages. I
> very rarely scroll before the page has completely loaded.
Further note: When NetSurf is first loaded I can scroll the first
page or two viewed with the scroll wheel without any problems. After
that it starts hanging and only moving one 'step' no matter how much I
scroll each time.
Seems I'm constantly updating my own thread...
Somewhere between r10684 and r10687 the problem has gone away. I've
browsed a dozen or more sites this evening since upgrading this
afternoon, many with long pages, and no instances of scroll wheel
initiated hangs! It's now fully responsive all the time. I'll keep
trying, but it looks like it may have 'gone away' or been fixed. Can
someone else who has experienced the problem give it a try as well?
Not sure how to update or close the bug report.
--
Grahame Parish
2:50 a.m.
On 11 Aug 2010 Grahame Parish <maillist.parish(a)millers-way.net>
wrote:
[snip]
Seems I'm constantly updating my own thread...
Somewhere between r10684 and r10687 the problem has gone away.
I've
browsed a dozen or more sites this evening since upgrading this
afternoon, many with long pages, and no instances of scroll wheel
initiated hangs! It's now fully responsive all the time. I'll keep
trying, but it looks like it may have 'gone away' or been fixed. Can
someone else who has experienced the problem give it a try as well?
Not sure how to update or close the bug report.
Yes, confirmed that scrolling seems to be normal here again; also, the
frequent hard disc access reported by others seems not to happen now.
Are these related?
RISC OS 5.16, FWIW.
With best wishes,
Peter.
--
Peter, \ / zfc Lu \ Prestbury, Cheltenham, Glos. GL52
Anne \/ __ __ \ England.
and / / \ | | |\ | / _ \ http://pnyoung.orpheusweb.co.uk
family / \__/ \_/ | \| \__/ \______________ pnyoung(a)ormail.co.uk
4:32 a.m.
On 12 Aug, Dr Peter Young wrote in message
<ab6f814551.pnyoung(a)pnyoung.ormail.co.uk>:
Yes, confirmed that scrolling seems to be normal here again; also,
the
frequent hard disc access reported by others seems not to happen now. Are
these related?
The SVN log entries for 10686 and 10687 could be relevant, perhaps? There's
likely to be less getting written to NetSurf's log file now, at least.
--
Steve Fryatt - Leeds, England
http://www.stevefryatt.org.uk/
7:34 a.m.
On 13 Aug 2010 Steve Fryatt <lists(a)stevefryatt.org.uk> wrote:
On 12 Aug, Dr Peter Young wrote in message
<ab6f814551.pnyoung(a)pnyoung.ormail.co.uk>:
> Yes, confirmed that scrolling seems to be normal here again;
also, the
> frequent hard disc access reported by others seems not to happen now. Are
> these related?
The SVN log entries for 10686 and 10687 could be relevant, perhaps?
There's
likely to be less getting written to NetSurf's log file now, at least.
Thanks for that. I do look at the SVN log entries from time to time,
but even on a good day I understand less than about 5% of them. I did
wonder about these two in this connection, but given my total
ignorance of C, I wouldn't really have known.
With best wishes,
Peter.
--
Peter, \ / zfc Lu \ Prestbury, Cheltenham, Glos. GL52
Anne \/ __ __ \ England.
and / / \ | | |\ | / _ \ http://pnyoung.orpheusweb.co.uk
family / \__/ \_/ | \| \__/ \______________ pnyoung(a)ormail.co.uk
9:11 a.m.
In article <f022564551.GrahameParish(a)grahame.parish>, Grahame Parish
<maillist.parish(a)millers-way.net> wrote:
In message <5fe1e53e51.GrahameParish(a)grahame.parish> Grahame
Parish
<maillist.parish(a)millers-way.net> wrote:
> In message <01ec473b51.GrahameParish(a)grahame.parish>
Grahame
> Parish <maillist.parish(a)millers-way.net> wrote:
Somewhere between r10684 and r10687 the problem has gone away.
I've browsed a dozen or more sites this evening since upgrading
this afternoon, many with long pages, and no instances of scroll
wheel initiated hangs! It's now fully responsive all the time.
I'll keep trying, but it looks like it may have 'gone away' or
been fixed. Can someone else who has experienced the problem give
it a try as well? Not sure how to update or close the bug report.
I think I asked before but have forgotten what the answer was - can
the scrolling be made faster as in EasiWriter and StrongED?
Richard
8:01 a.m.
In article <5145a44f1driscos(a)gotadsl.co.uk>,
Richard Ashbery <riscos(a)gotadsl.co.uk> wrote:
I think I asked before but have forgotten what the answer was - can
the scrolling be made faster as in EasiWriter and StrongED?
Not sure what you mean. If you mean scrolling with a mouse scrollwheel,
then you can configure how fast it goes on a per-application basis in HID.
If you mean the general redraw speed (e.g. when dragging the scrollbar),
then there is little that can be done to speed it up. It's already fast
here. Are there any pages that scroll particularly slowly?
--
Michael Drake (tlsa) http://www.netsurf-browser.org/
7:15 a.m.
In article <5147296f0dtlsa(a)netsurf-browser.org>, Michael Drake
<tlsa(a)netsurf-browser.org> wrote:
In article <5145a44f1driscos(a)gotadsl.co.uk>, Richard Ashbery
<riscos(a)gotadsl.co.uk> wrote:
> I think I asked before but have forgotten what the answer was -
> can the scrolling be made faster as in EasiWriter and StrongED?
Not sure what you mean. If you mean scrolling with a mouse
scrollwheel, then you can configure how fast it goes on a
per-application basis in HID.
I do mean scrolling with the mouse scrollwheel - unfortunately I have
never been able to get HID to work with my Logitech LX3. As soon as I
run HID the pointer disappears so I am unable to configure it. Has
anyone else had this problem?
If you mean the general redraw speed (e.g. when dragging the
scrollbar), then there is little that can be done to speed it up.
It's already fast here. Are there any pages that scroll
particularly slowly?
The amount the scroll bar moves in one physical movement of the mouse
scrollwheel is the problem not the redraw speed.
This is not a scientific test but hopefully will illustrate what I
mean. Load the following webpage......
http://www.riscos.info/index.php/MassFS_OtherDevs_entries.
Copy the page into a StrongED text document and place side by side on
the screen.
Start at the top of the NetSurf page and use the scroll wheel to
scroll down to the bottom. Count the number of times you have to move
the scroll wheel to get to the bottom of the page. Do the same for the
StrongED document. In NetSurf it takes about 23 wheel movements to do
this but in StrongED it takes only 5.
Richard
(Using RISC OS 5.13. I am going to update but so many other things get
in the way)
3:56 p.m.
In article <1281958995.4c69245373460(a)netmail.pipex.net>, dave higton
<davehigton(a)dsl.pipex.com> wrote:
Quoting Richard Ashbery <riscos(a)gotadsl.co.uk>:
> (Using RISC OS 5.13. I am going to update but so many other
> things get
Richard, you really, really, really need to upgrade to 5.16.
The process is quick and easy. The longest part is the downloading
of the file.
Thanks for the reminder Dave - I will as soon as possible.
Richard
4672
days inactive
4700
days old
netsurf-users@netsurf-browser.org
23 comments
7 participants
participants (7)
-
dave higton -
Dr Peter Young -
Grahame Parish -
Michael Drake -
Richard Ashbery -
Richard Porter -
Steve Fryatt